Raster Calculation and Weighted Sum

[deleted-user-wXI1z2K2eX3q]

I am attempting to create a cost surface that incorporates the use of a DEM (slope raster), a Landcover Raster, Strahler stream order raster, lakes raster, roads raster, and a trails raster. I am under the assumption that the traveler will ALWAYS choose the trail or the road. I have many start and end points to run my least cost paths. I have also already reclassified all the rasters to a common scale. 

I would like to know if there is anyone who knows how it is possible for me to somehow burn the roads and trails rasters into the other rasters. I would like the trails and roads to retain a value of "1" after raster calculation or weighted sum, while all other raster values are added together. Is this possible without weighting each raster differently?

[DanPatterson_Retired]

you can obviously just use the Con tool  if the road layer exists on its own, and your cost surface is another layer, it would be in the form

Con("Road" == 1, 1, "CostSurface")  assuming your roads are classed as 1 and non-road 0 

Check the examples in the help topics and on GeoNet, there are ton of 'Con' examples

[deleted-user-wXI1z2K2eX3q]

Obviously. Thank you, Dan.

[DanPatterson_Retired]

Raymond, I meant 'obviously' in that there is so little in the strictly 'conditional' toolset to choose from.

Con should be renamed 'Where'

# ---- make a cost surface -----
costsurface = np.random.randint(2, 20, size=(10, 10))

# ---- make some roads ----
roads = np.zeros_like(costsurface)
roads[:, 2] = 1
roads[2, :] = 1
roads 
array([[0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 0, 0, 0, 0, 0]])

# ---- Find out 'where' the roads are in the cost surface ---
# and 'embed' into it
embedded = np.where(roads == 1, roads, costsurface)

embedded
array([[ 6,  3,  1, 17, 16,  8,  5,  5,  2, 13],
       [13,  3,  1,  7,  6, 10,  7, 17,  5, 15],
       [ 1,  1,  1,  1,  1,  1,  1,  1,  1,  1],
       [14,  6,  1,  6, 10, 14, 15,  5,  9,  7],
       [ 4, 18,  1, 13, 11, 12, 16, 15,  3, 12],
       [ 7,  3,  1, 14,  3, 15, 19, 19, 12, 12],
       [ 7, 11,  1, 19,  7,  4, 13,  2, 11,  8],
       [17, 19,  1,  9,  9, 14, 14, 15, 13,  5],
       [11,  2,  1, 15,  8,  3, 15,  4, 19, 11],
       [17,  4,  1,  7,  2,  7,  7, 12, 12,  9]])
‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

[deleted-user-wXI1z2K2eX3q]

No worries! I must have read your reply during a state of frustration . I really appreciate your swift and accurate response! Con tool worked beautifully.

[DanPatterson_Retired]

Glad it worked out Raymond